Diana Romero

Immunotherapy: PD-1 says goodbye, TIM-3 says hello

notably antibodies targeting the PD-1/PD-L1 axis, have demonstrated efficacy as therapeutic agents for several tumour types. As Peter Hammerman explains, however, “little is known about which patients are likely to respond to these agents and, if they respond, the possible mechanisms of resistance to therapy are unknown.” Findings now published by Hammerman and colleagues show that TIM-3 and other immune checkpoints are upregulated as a result of adaptative resistance. In the first part of this study, the investigators used two genetic models of lung cancer driven by mutations in genes that are clinically relevant to this malignancy, such as EGFR and KRAS, in fully immunocompetent mice. The tumour immune microenvironment was analysed before, during and after treatment with a PD-1-blocking antibody. The expression of several immune checkpoints — most notably, TIM-3 — was upregulated in T cells from animals that developed resistance to the anti-PD-1 treatment. In independent experiments, resistance to anti-PD-1 therapy was prevented when an anti-TIM-3 antibody was administered together with an anti-PD-1 agent. Importantly, the authors undertook validation studies using samples derived from two patients with lung adenocarcinoma who had developed progressive disease after receiving anti-PD-1 therapy. The immune cells present in effusion samples collected from these patients were analysed, and their phenotypic profile was compared with that of immune cells from five patients with non-smallcell lung cancer who had not received anti-PD-1 treatment; higher levels of TIM-3, but not of other immune markers, were detected in T cells from the patients who had developed resistance to anti-PD-1 therapy compared with the other patients. These results indicate that, in the setting of anti-PD-1 therapy, treatment failure is associated with upregulation of alternative immune checkpoints that act to limit the antitumour immune response. Hammerman indicates, “this mechanism had been demonstrated in the infectious disease literature, but this is the first report in a cancer context”. Future studies need to address how the immune response can be monitored during treatment in order to identify novel approaches to potentiate the effect of PD-1/PD-L1 blockade. As Hammerman notes, “responses to PD-1/PD-L1 therapy remain suboptimal in the majority of patients and there is much to learn and improve on.” Diana Romero I M M U N OT H E R A P Y

Breast cancer: Tracking ctDNA to evaluate relapse risk

Identifying the patients who are most at risk of relapse after treatment of primary breast cancer is a major challenge. Circulating tumour DNA (ctDNA) present in the blood of patients with breast cancer can be used to determine whether minimal residual disease (MRD) persists after treatment, according to findings of a recent study led by Nicholas Turner. The 55 patients involved in this study received neoadjuvant chemotherapy. DNA samples obtained before any treatment was initiated were screened for breastcancer-related somatic mutations. These mutations were subsequently tracked in postsurgical blood samples using digital PCR, to enable discrimination of ctDNA from other circulating cell-free DNA. ctDNA was analyzed at different time points after surgery. For 80% of the patients who had a disease relapse, ctDNA was detected in serial samples; for 50% of them, ctDNA was detected in a single postsurgical sample. The prediction of BREAST CANCER

Haematological cancer: MRD assessment - guiding decisions for patients with AML.

leukaemia (AML) in young adult patients has not changed much over the past 30 years. The risk of relapse can be reduced in patients who receive a transplant. “AML is a heterogeneous disease and we have very limited information to predict how likely a patient is to benefit from a transplant”, says David Grimwade. Grimwade is the principal investigator of a new study, results of which show that assessing minimal residual disease (MRD) in blood has prognostic value for risk of relapse and, thus, can inform treatment decisions. RT-qPCR was performed in samples from 346 patients with mutations in NPM1 (present in one-third of patients with AML) who had participated in the UK National Cancer Research Institute AML17 trial of therapies for AML. The persistence of NPM1-mutated transcripts in peripheral blood following the second cycle of chemotherapy was indicative of slow clearance of AML. Importantly, patients with MRD had a higher 3-year relapse risk than those patients in whom the transcripts were absent (HR 4.80; CI: 2.95–7.80; P <0.001). These findings were validated in a separate cohort of 91 patients from the AML17 trial. These results indicate that MRD can be assessed with RT-qPCR in peripheral blood. This is “a clean and very powerful method,” in Grimwade’s words, to predict at an early point whether patients with AML are responding to therapy, and thus could help guide subsequent treatment decisions. Grimwade and collaborators are currently conducting a trial to further assess how, using this method of genetic tracking, “we will be able to avoid doing a transplant in patients that do not need it because they will not benefit from it.” Diana Romero H A E M ATO LO G I C A L C A N C E R

Haematological cancer: ETV6 germline mutation - a risk for ALL.

generally not considerered as a risk factor for the development of acute lymphoblastic leukaemia (ALL), the most-common cancer in children. Now, a study led by Jun Yang, shows that the presence of a familial component in ALL pathogenesis is much greater than it was previously believed. Yang and colleagues identified a family with an unusual clustering of childhood ALL. With the hypothesis that this family had a genetic predisposition to developing ALL, whole-exome-sequencing analysis was performed for all family members. Among the nonsilent variants identified in patients with ALL from this family, priority was given to a novel ETV6 variant. The reason for focusing on this variant was that somatic mutations in ETV6 have already been linked to the development of haematological malignancies. To confirm this association, targeted next-generation sequencing of ETV6 was performed in germline DNA samples from 4,405 children with ALL. In this large cohort, 31 rare ETV6 variants were identified that were potentially related to ALL predisposition. In children harbouring these mutations, ALL had distinct clinical features, such as older age at diagnosis, which indicate a unique mechanism of leukaemia pathogenesis related to these ETV6 variants. The results of this study revealed that, although rare, the genetic predisposition for ALL does not occur only in exceptional cases. The authors estimate that approximately 1% of patients with childhood ALL potentially carry highly penetrant ETV6 variants. These findings provide a strong rationale for including family history examination as part of the standard approach to treating paediatric ALL. As Yang indicates, “guidelines are needed for the clinical interventions and surveillance of individuals with inherited predisposition variants.”

Lung cancer: Nivolumab[mdash]an effective second-line treatment for NSCLC

Among the 582 patients enrolled in this trial, 292 were treated with nivolumab and 290 received docetaxel. The primary end point of the trial was overall survival; secondary end points included the objective response rate, progressionfree survival (PFS), efficacy according to tumour PD-L1 expression level, and patient-reported outcomes. Nivolumab was associated with a significant survival benefit over docetaxel, with a median overall survival of 12.2 months versus 9.4 months. The survival benefit was observed at different time points, with overall survival rates of 51% for nivolumab versus 39% for docetaxel at 1 year, and 39% versus 23%, respectively, at 18 months. The objective response rate was also significantly higher for patients treated with nivolumab (19%) compared with those in the docetaxel group (12%). Interestingly, the median PFS duration was 2.3 months with nivolumab and 4.2 months with docetaxel, but a clear LUNG CANCER

To all involved — we have a problem

© 2018 Macmillan Publishers Limited, part of Springer Nature. All rights reserved. Tempting as it is to adopt a simplistic wellmeaning approach, we don’t consider that merely publishing articles on the financial toxicity of cancer constitutes an achievement for Nature Reviews Clinical Oncology as a journal for the medical oncology community. But we regularly feature articles on the cost of anticancer treatments and have been doing so for more than a decade. What made us consider those articles appropriate for our journal is the proactive approach of their authors, which we hope our readers have also found inspiring. In a nutshell: we publish discussions of potential solutions and avoid ‘moaning exercises’ (empty descriptions of the current bleak panorama). Two perfect examples of the type of article we consider of interest for our readers are featured in the July 2018 issue of the journal. In a provocative Comment, Vinay Prasad, Christopher McCabe and Sham Mailankody discuss a “thought experiment” (ref.1) in which they evaluate whether pharmaceutical companies could derive profits from conducting clinical trials with inert anticancer therapies. Their conclusion is sobering: any anticancer drug that generates a US

Breast cancer: Metronomic chemotherapy for elderly and/or frail patients

440 million profit and is approved on the basis of the results of a single clinical trial would justify a hypothetical portfolio involving 100 inert compounds. This scenario is an exaggerated distortion of reality, and the authors “certainly do not believe that companies are actively pursuing ineffective drugs”, but the revenue they propose actually matches those of many agents currently used in clinical practice. Let’s not forget that anticancer agents remain the bestselling drugs among FDAapproved therapies (32% of sales projected in 2017)2. Another fact to keep in mind is that many drug approvals are indeed based on the results of a single trial, which do not always meet the threshold for meaningful clinical benefit3. Mailankody et al. state that the risk– benefit balance in oncology clinical trials (regardless of whether they are intended to lead to drug approval) remains to be properly addressed. They do not explicitly formulate a request but, after reading their article, we cannot help but ask for transparency from the regulatory bodies regarding the criteria they use for drug approvals. Apropos of transparency, Carin Uylde Groot and Bob Löwenberg believe that this is the principle that should guide the pricing of anticancer agents. In their News & Views article4 they present an algorithm that combines the key factors that should be considered to set the price of a drug: R&D costs, marketrelated costs, patent duration, target patient population and profit margin. Of note, in their model, profit is proportionally related to clinical benefit. These authors directly address the role of regulators who, in their opinion, should actively guarantee that anticancer agents are fairly priced and promote access to them. Clinical benefit is a common theme in these two articles. ‘Value’ and ‘benefit’ can mean different things, even in the context of patient management. To avoid ambiguity, a few years ago several groups of experts independently established frameworks to define clinical benefit through a series of set criteria. Importantly, this commendable effort was led by the two major medical oncology societies, ESMO and ASCO. As proposed, we consider that regulatory bodies should ensure transparency in drug approvals and in the subsequent pricing of such drugs. Perhaps only societies with the reputation of ESMO and ASCO will be able to reach out to the regulatory bodies and ask them to bridge the gap between what their frameworks define as meaningful clinical benefit and the outcomes derived from treatment with most of the innovative agents currently available. Both articles have been written by professionals from an academic environment and located in highincome countries. As editors of Nature Reviews Clinical Oncology, we will keep publishing articles about value if, by doing so, we can foster real discussion. We are aware that we need to include all the voices that belong in this debate and have a constructive opinion. We are interested, in particular, in hearing from professionals in the pharmaceutical industry. We understand that drug development needs to be profitable to those who invest an effort in such a complex task. We do not have clear information, however, about the costs associated with the different stages of marketing of a therapeutic agent, from the manufacturer to the doctor’s office. We have a problem: the rising cost of anticancer therapies and the current regulatory environment have helped to create an unsustainable (and unacceptable) situ ation. This problem frequently receives attention in our journal — and we consider that our duty is to do so. This is a call to all stakeholders to state what their position is in face of this problem. We, as editors, are committed to open the communication channel to everyone with a meaningful contribution. Changes will not happen quickly, but they need to happen.

Chemotherapy: Rolapitant—a new and safer antiemetic agent

Hans Wildiers show that, for older patients with metastatic HER2-positive disease, the addition of metronomic cyclophosphamide to the combination trastuzumab plus pertuzumab (TP) leads to markedly longer progression-free survival (PFS) durations compared with TP alone. The investigators chose metronomic chemotherapy because of its lower risk of toxicities compared with docetaxel, which is usually administered in combination with TP. “The goals of treatment become different with increasing age: overall survival is less relevant in older women than in younger patients. Avoidance of classical chemotherapy is an important goal for the elderly,” Wildiers explains. Patients in this study had chemotherapynaive metastatic HER2-positive breast cancer and were either at least 70 years old, or ≥60 years old and functionally impaired as defined by a series of predefined criteria. Participants were randomly assigned to receive TP alone (n = 39) or supplemented with metronomic oral cyclophosphamide (TPM; n = 41). At 20.7 months, patients receiving TPM had longer PFS durations than those receiving TP (12.7 months versus 5.6 months). Patients with disease progression during treatment were eligible for subsequent monotherapy with trastuzumab entamsine. This treatment, which was given to 29 patients in this study, resulted in median disease-control durations of 5 months. Of note, all of the regimens tested had acceptable toxicity profiles. “The benefit of avoiding the toxicity of chemotherapy by administering TP only does not compensate for a substantial loss of activity. Antitumour activity remains higher when a ‘gentle’ chemotherapy is added to TP,” Wildiers concludes. Diana Romero B R E A S T C A N C E R

Haematological cancer: promising results of BCL2 inhibition.

Chemotherapy-induced nausea and vomiting (CINV) is one of the most feared adverse effects of chemotherapy. Now, three trials, all coordinated by Lee Schwartzberg and Bernardo Rapoport, show that rolapitant is a novel, safe and effective agent for the prevention of CINV. CINV is divided into an acute phase (the first 24 h after treatment initiation) and a delayed phase (24–120 h after treatment); serotonin is the main mediator of the acute phase reactions, whereas those that occur in the delayed phase are mediated by neurokinin-1 (NK-1) and substance P. Thus, current treatment guidelines for the prevention of CINV recommend the use of an NK-1 receptor antagonist, a serotonin receptor antagonist and dexamethasone. However, most NK-1 receptor antagonists interact with CYP3A4 and, therefore, affect the pharmacokinetics of drugs administered concomitantly. Rolapitant is an NK-1 receptor antagonist with high affinity for the human NK-1 receptor that does not interact with CYP3A4. Its half-life is around 180 h and thus a single dose might be sufficient to prevent CINV. The aim of three phase III trials was to assess the safety and efficacy of rolapitant in combination with granisetron (a 5-HT3 receptor antagonist) and dexamethasone to prevent CINV induced by moderately emetic chemotherapy (MEC) and highly emetic chemotherapy (HEC). The MEC agents used were carboplatin, cyclophosphamide and irinotecan, among others. The HEC agents used were cisplatin, and anthracycline plus cyclophosphamide combinations. In both the MEC and HEC settings, a significantly greater proportion of patients treated with rolapitant compared with placebo had no delayed-phase emesis, which addresses an unmet clinical need. In addition, patients receiving rolapitant who did not develop CINV in the acute phase were more likely to not develop CINV in the delayed phase. “Rolapitant’s long half-life offers us the opportunity to further explore dosing options—maybe it can be given the day before chemotherapy if that is more convenient for the patient,” says Rapoport. Schwartzberg concludes, “patients often believe that they are supposed to vomit on chemotherapy, but the reality is we have very good drugs to prevent this. Some patients will stop their treatment because of their CINV and that’s the last thing we want.”

Expanding ribociclib use

Although often associated with a favorable prognosis, acute lymphoblastic leukaemia (ALL) is sometimes refractory to treatment. Studies have shown that the TCF3–HLF fusion transcript defines a group of patients with ALL who have a particularly poor prognosis. Using an integrated drugresponse profiling approach in patientderived xenografts, an international collaborative group lead by Martin Stanulla and Jean-Pierre Bourquin identified alternative treatment options for this deadly disease. Combining genomic and functional studies to further profile TCF3–HLFpositive ALL, they found that the TCF3–HLF translocation occurs only in B-lymphoid progenitor cells and is accompanied by genetic alterations in PAX5, BTG1 or VPREB1 that impair pro-B cell differentiation into pre-B cells. Comparison with samples containing the TCF3–PBX1 fusion, which is associated with a better prognosis, indicated that although both fusions occur in B-lymphoid cells, TCF3–HLF promotes transcriptional reprogramming towards a drug-resistant very immature state. The researchers generated leukaemia xenografts for all the samples included in this study. The molecular features in these xenografts closely resembled the original patient samples. Drug-activity profiling in the xenografts revealed that TCF3–HLF ALL has a unique sensitivity to the BCL2targeting agent venetoclax. Combination of this drug with conventional antileukaemic therapy induced prolonged remissions in patient-derived xenografts. This study describes a valuable preclinical model to evaluate different treatment strategies for patients with ALL, which could provide a solid rationale for clinical trials. On the basis of these results, translational scientists at different centres across the world have proposed a joint effort to test venetoclax as part of a combination therapy for several ALL subtypes. As Bourquin notes, “this effort would help identify the right group of patients for treatment with the drug combinations that are right for them.” Importantly, he also states that, “it would be up to the pharmaceutical industry to take the challenge and collaborate with us in initiatives that will have a great impact for patients.”